探讨术前血小板-白蛋白（PAL）评分对肝硬化肝细胞癌（肝癌）患者术后预测价值。

回顾性分析2016年1月至2022年1月在川北医学院附属医院行肝切除术的171例肝硬化肝癌患者临床资料。其中男145例，女26例；年龄24~78岁，中位年龄50岁。患者均合并乙型病毒性肝炎。患者均签署知情同意书，符合医学伦理学规定。根据PAL评分患者分为PALⅠ组（101例）和PALⅡ-Ⅲ组（70例）。两组术前PAL评分比较采用t检验。两组肝切除术后肝衰竭（PHLF）发生率、围手术期死亡率比较采用χ²检验。采用ROC曲线分析PAL评分对PHLF、围手术期死亡的预测价值。生存分析采用Kaplan-Meier法和Log-Rank检验。

PALⅠ组术前PAL评分为（-4.07±0.20）分，明显低于PALⅡ-Ⅲ组的（-3.45±0.31）分（t=-15.589，P<0.05）。PALⅠ、Ⅱ、Ⅲ级患者发生PHLF分别为9、15、7例，围手术期死亡相应为1、5、2例。PALⅡ-Ⅲ组PHLF发生率为31%（22/70），明显高于PALⅠ组的9%（9/101） （χ²=14.125，P<0.05）；围手术期死亡率为10%（7/70），亦明显高于PALⅠ组的1%（1/101） （χ²=5.641，P<0.05）；随着PAL等级越高，发生PHLF的风险越大，围手术期死亡率也越高（χ²=23.246, 9.485；P<0.05）。PAL评分预测PHLF和围手术期死亡的ROC曲线下面积（AUC）分别为0.712、0.768，Child-Pugh分级预测相应为0.617、0.645，两个模型预测PHLF和围手术期死亡的AUC差异无统计学意义（Z=1.380，1.235；P>0.05）。PALⅠ组术后1、3、5年累积生存率分别为84.6%、55.6%、34.6%，PALⅡ-Ⅲ组相应为75.2%、40.5%、27.7%，PALⅠ组术后总体生存优于PALⅡ-Ⅲ组（χ²=5.698，P<0.05）。

PAL评分对肝硬化肝癌患者术后肝衰竭、围手术期死亡及远期预后有一定的预测价值，具有与Child-Pugh分级系统相当的效能。

To evaluate the predictive value of preoperative platelet-albumin (PAL) score for the postoperative prognosis of patients with cirrhotic hepatocellular carcinoma (HCC).

Clinical data of 171 HCC patients with cirrhotic HCC who underwent hepatectomy in the Affiliated Hospital of North Sichuan Medical College from January, 2016 to January, 2022 were retrospectively analyzed. Among them, 145 patients were male and 26 female, aged from 24 to 78 years, with a median age of 50 years. All patients were complicated with hepatitis B. The informed consents of all patients were obtained and the local ethical committee approval was received. According to PAL score, all patients were divided into the PALⅠ group (n=101) and PAL Ⅱ-Ⅲ group (n=70). Preoperative PAL scores between two groups were compared by t test. The incidence of post-hepatectomy liver failure (PHLF) and perioperative mortality between two groups were compared by using Chi-square test. The predictive value of PAL score for PHLF and perioperative mortality was analyzed by the ROC curve. Survival analysis was performed by Kaplan-Meier method and log-rank test.

In the PALⅠ group, preoperative PAL score was -4.07±0.20, significantly lower than -3.45±0.31 in the PAL Ⅱ-Ⅲ group (t=-15.589, P<0.05). In the PALⅠ, Ⅱ and Ⅲ patients, 9, 15 and 7 cases developed PHLF, respectively, and 1, 5 and 2 died during perioperative period. The incidence rate of PHLF in the PAL Ⅱ-Ⅲ group was 31%(22/70), significantly higher than 9%(9/101) in the PAL Ⅰ group (χ²=14.125, P<0.05). Perioperative mortality was 10%(7/70) in the PAL Ⅱ-Ⅲ group, significantly higher than 1%(1/101) in the PAL Ⅰ group (χ²=5.641, P<0.05). The higher the PAL grade, the higher the risk of PHLF and the higher the perioperative mortality (χ²=23.246, 9.485; P<0.05). The area under ROC curve (AUC) of PAL score in predicting PHLF and perioperative mortality was 0.712 and 0.768, and 0.617 and 0.645 for Child-Pugh score, respectively. No significant difference was observed in the AUC between two models for predicting PHLF and perioperative mortality (Z=1.380, 1.235; P>0.05). In the PALⅠ group, the postoperative 1-, 3- and 5-year cumulative survival rates were 84.6%, 55.6% and 34.6%, and 75.2%, 40.5% and 27.7% in the PAL Ⅱ-Ⅲ group, accordingly. The postoperative overall survival in the PALⅠ group was better than that in the PAL Ⅱ-Ⅲ group(χ²=5.698, P<0.05).

PAL score has certain predictive value for PHLF, perioperative mortality and long-term prognosis of patients with cirrhotic HCC, which yields equivalent efficiency to Child-Pugh classification.